Heterojunction-Based Photocatalytic Degradation of Rose Bengal Dye via Gold-Decorated α-Fe₂O₃-CeO₂ Nanocomposites under Visible-Light Irradiation

Abstract

Developing photocatalytic nanomaterials with unique physical and chemical features using low-cost and eco-friendly synthetic methods is highly desirable in wastewater treatment. In this work, the magnetically separable α-Fe₂O₃-CeO₂ nanocomposite (NC), with its respective metal oxides of α-Fe₂O₃ and CeO₂ nanoparticles, was synthesized using a combination of hexadecyltrimethylammonium bromide (CATB) and ascorbic acid via the hydrothermal method. To tune the band gap, the heterojunction nanocomposite of α-Fe₂O₃-CeO₂ was decorated with plasmonic Au nanoparticles (Au NPs). The various characterization methods, such as FTIR, UV-vis DRS, XRD, XPS, TEM, EDX, SEM, and PL, were used to determine the properties of the materials, including their morphology, elemental composition, optical properties, band gap energy, and crystalline phase. The nanocomposite of α-Fe₂O₃-CeO₂@Au was utilized to remove Rose Bengal (RB) dye from wastewater using a photocatalytic technique when exposed to visible light. A comprehensive investigation of the impact of the catalyst concentration and initial dye concentration was conducted to establish the optimal photodegradation conditions. The maximum photocatalytic efficiency of α-Fe₂O₃-CeO₂@Au (50 mg L⁻¹) for RB (20 ppm) dye removal was found to be 88.9% in 120 min under visible-light irradiation at a neutral pH of 7 and 30 °C. Various scavengers, such as benzoquinone (BQ; 0.5 mM), tert-butyl alcohol (TBA; 0.5 mM), and ethylenediaminetetraacetic acid (EDTA; 0.5 mM), were used to investigate the effects of different free radicals on the photocatalytic process. Furthermore, the reusability of the α-Fe₂O₃-CeO₂@Au photocatalyst has also been explored. Furthermore, the investigation of the potential mechanism demonstrated that the heterojunction formed between α-Fe₂O₃ and CeO₂, in combination with the presence of deposited Au NPs, led to an enhanced photocatalytic efficiency by effectively separating the photogenerated electron (e⁻)–hole (h⁺) pairs.

1. Introduction

The production of synthetic dyes, such as fabrics, papers, cosmetics, and leather, has increased in several industries [1,2]. Approximately 7 × 10⁷ tons of these toxic dyes are manufactured annually [3], and often 5 × 10³ tons are released into the aquatic environment as wastewater [4], posing a threat to human health and other living organisms [5,6]. One such dye is Rose Bengal (4,5,6,7-tetrachloro-2′,4′,5′,7′-tetraiodofluorescein), also recognized as Acid Red 94, a synthetic dye belonging to the basic xanthene group. Rose Bengal is water-soluble and comprises functional groups such as hydroxyl and carboxyl, along with aromatic rings crucial for its molecular interactions. It is widely employed in various industries, like textiles, cosmetics, medical diagnostics, and microbiology, as a staining agent. However, its widespread use has resulted in its presence in wastewater effluents. Despite its utility, this dye is hazardous and capable of causing skin and eye irritation. If released into water bodies, it can pose a significant threat to aquatic ecosystems. Furthermore, it is not biodegradable, which makes it necessary to have effective strategies for its disposal and removal from wastewater. Fortunately, Rose Bengal is light-sensitive, particularly to visible and UV wavelengths, rendering it prone to photodegradation [7,8,9,10,11,12].

Researchers have tried to remove pollutants from wastewater using conventional methods such as adsorption [13,14,15,16], oxidation [17], biodegradation [18], and membrane [19], but they are both inefficient and expensive [8]. Therefore, there is a crucial demand for an affordable and environmentally friendly approach to treating dye wastewater before it is disposed of into the environment [19,20,21,22]. Photocatalysis is mainly used to degrade textile dyes using effective semiconductor photocatalysts [22,23]. The limited recombination rate of electron–hole pairs are responsible for efficient degradation upon solar or UV–visible-light excitation [21,22,23]. Semiconductor nanomaterials ranging from 1 to 100 nm have garnered significant interest in photocatalysis because of their distinct physical and chemical properties compared to bulk materials [22,23,24]. Among others, transition metal oxide and semiconductor nanomaterials are appealing candidates in the photolysis field [21,22,23] because of their biocompatibility, remarkable stability, and capacity to produce charge carriers when stimulated with sufficient light energy [23,24].

Chemical catalysis and energy conversion are two areas where intelligent and effective nanomaterials are gaining popularity. Nanostructured metal oxides, including CeO₂, Fe₂O₃, TiO₂, MnO₂, ZnO, Fe₃O₄, and SnO₂, have been used to drive chemical reactions instead of conventional energies [25,26,27,28,29,30,31]. Of these metal oxides, α-Fe₂O₃ is a semiconductor with great potential for the breakdown of organic contaminants through photocatalysis and the generation of hydrogen in water splitting because of its natural abundance, chemical stability, low band gap, and environmental friendliness [32,33]. Although possessing these characteristics, α-Fe₂O₃ exhibits a high recombination rate due to the short path length for hole diffusion and its restricted oxidation capability. This rapid recombination of electron–hole pairs hinders the commercial viability of the material. To address this issue, it is essential to incorporate additional active materials with a greater surface area in the host matrix [34].

Many researchers have created heterojunction composites, including α-Fe₂O₃@TiO₂, α-Fe₂O₃@ZnO, Fe₃O₄-ZnO/TiO₂, α-Fe₂O₃@SnO₂, Fe₃O₄-CdO, and TiO₂-CeO₂, in an attempt to harvest solar energy, reduce the charge carrier recombination rate, and increase the photocatalytic activity [35,36,37,38,39,40]. To improve electron–hole pair separation, rare-earth oxides can be combined with α-Fe₂O₃ [41]. Because of its ability to both gain and lose oxygen, the multifunctional rare-earth metal oxide CeO₂ is one of several oxides of rare-earth metals that could be used in photocatalysis. The narrow band gap and oxidative–reductive capability of the Ce(III)/Ce(IV) pair make CeO₂ an attractive photocatalytic candidate [42]. An increase in photocatalytic activity can be achieved through the reversible transformation of Ce(III)/Ce(IV) couples, which allows for the movement of electrons and the diffusion of charge carriers between CeO₂ and other semiconductors, like TiO₂, MoS₂, and CdS [43,44,45]. Thus, creating a heterojunction of Fe₂O₃ with CeO₂ makes sense, has also enhanced the redox stability of the material, and is essential for photocatalytic performance. Numerous scientists have developed heterojunction composites comprising Fe₂O₃ and CeO₂ to explore their potential in various applications, such as photocatalytic hydrogen evolution, ethanol conversion, supercapacitor technology, and dye degradation [46,47,48,49].

Moreover, researchers have observed that covering the surface of NPs with noble metals like platinum, silver, or gold can further enhance their electron–hole separation and photocatalytic activity [50,51,52]. The surface plasmon resonance (SPR) effects of these nanoparticles can considerably promote the ability of photocatalysts to respond to visible light [50,51,52]. Gold nanoparticles, in particular, are highly stable, with enhanced visible-light absorption and efficient charge separation, and can serve as an electron intermediary during charge transfer, significantly preventing the recombination of photogenerated electron–hole pairs [53,54].

In this study, a hydrothermal method was used to fabricate CeO₂, α-Fe₂O₃, and the heterojunction of α-Fe₂O₃-CeO₂ nanoparticles combining different materials (CTAB and ascorbic acid) for surface modification to achieve improved physical and chemical properties [55,56]. CTAB is frequently utilized as a template to enhance uniformity and create a mesoporous structure [57,58], while ascorbic acid, with its four hydroxyl groups, functions as a ligand for metal oxide nanoparticles [59]. The visible-light absorption and decreased e⁻-h+ pair recombination improved with the decoration of Au NPs on the surface of α-Fe₂O₃-CeO₂ NC.

Gold ions were effectively transformed into nanoparticles by utilizing jujube extract as a reducing agent [60,61,62]. The α-Fe₂O₃-CeO₂@Au nanocomposite was designed to degrade RB dye, aiming to minimize the production of harmful byproducts and secondary pollution, in line with the sustainability goals [63]. The degradation was achieved under visible light. This light is a significant portion of solar radiation, making it a great source of renewable energy and reducing energy consumption compared to UV light [64]. The reusability capabilities of the as-synthesized photocatalyst were investigated to remove RB. A proposed mechanism for a photocatalytic process driven by visible light was also presented.

2. Materials and Methods

2.1. Materials

The reagents used in this study include iron(III) nitrate nonahydrate (Fe₃(NO₃)₃.9H₂O, purity 98.0%), cerium(III) nitrate (Ce(NO₃)₃. 6H₂O, purity 99.0%), Hexadecyltrimethylammonium bromide (C₁₉H₄₂BrN; purity 99.0%), L-ascorbic acid (C₆H₈O₆; purity 99.0%), Ethanol, (C₂H₅OH; Purity 99.5%), Ammonia (NH₃; purity 99.0%), Gold(III) chloride (AuCl₃; purity 99.0%), and Rose Bengal dye (C₂₀H₂Cl₄I₄Na₂O₅; purity 95.0%) were acquired from Sigma-Aldrich, St. Louis, MO, USA. Ziziphus jujuba fruit was purchased from the local market. All chemicals were utilized as received without additional purification. All of the solutions were prepared using deionized water.

2.2. Synthesis of Fe₂O₃, CeO₂, Fe₂O₃-CeO₂ and Fe₂O₃-CeO₂@Au NPs

For the preparation of Fe₂O₃ and CeO₂ NPs, a hydrothermal approach was carried out [65,66]. Initially, 2 g of Fe(NO₃)₃·9H₂O and Ce(NO₃)₃.·6H₂O were separately dissolved in 50 mL of deionized water to create a homogeneous solution. An aqueous solution of CTAB (10 mL, 0.2 M) and ascorbic acid (20 mL, 1.0 M) was added to each metal ion solution with continuous stirring at 50 °C. After that, aqueous ammonia was slowly added, drop by drop, to the reaction mixture, until a pH of 12 was achieved [67,68]. The mixture was moved into a Teflon-lined autoclave with a volume of 100 mL. Following that, it was subjected to heating at 180 °C in the oven for a period of 24 h. The obtained material was centrifuged at 4000 rpm for 20 min and rinsed multiple times with deionized water and ethanol. The obtained solid samples were subjected to drying at a temperature of 70 °C for a duration of 24 h, and subsequently subjected to calcination at a temperature of 500 °C for a duration of four hours.

Furthermore, α-Fe₂O₃-CeO₂ NC was prepared by mixing an equal quantity of metal ion solutions of Fe(NO₃)₃·9H₂O and Ce(NO₃)₃·6H₂O in one beaker and an aqueous solution of 10 mL CTAB (0.2 M) and 20 mL ascorbic acid (1.0 M) were added with continuous stirring at 50 °C. Aqueous ammonia was slowly added, drop by drop, to this mixture until the pH reached approximately 12. The mixture was moved into a Teflon-lined autoclave and heated at 180 °C in the oven for a period of 24 h. After allowing the autoclave to cool down, the as-prepared material was collected by centrifugation for 20 min at 4000 rpm and subsequently washed with deionized water and ethanol, followed by drying at 70 °C for 24 h and calcination at 500 °C for 4 h.

All prepared samples were kept in an airtight, dark place for additional characterization and use. The Au-supported α-Fe₂O₃-CeO₂ NPs were prepared by facile green synthesis technique employing aqueous jujube fruit extract as both a reducing and capping agent [69,70]. Initially, 20 g of dried jujube fruit was washed several times with deionized water and then air-dried at room temperature. The completely dried jujube fruit was crushed into fine powder and boiled for 20 min in 250 mL of deionized water. After being cooled down, the aqueous extract was filtered and then kept in a refrigerator for future utilization. The Au NPs deposition on the surface of α-Fe₂O₃-CeO₂ NC was achieved by dispersing 1.0 g of α-Fe₂O₃-CeO₂ NC in 10 mL of deionized water and vigorously stirring for 30 min. To this suspension, 15 mL of 0.20 M AuCl₃ aqueous solution was added and, subsequently, 10 mL of the jujube aqueous extract was added with continuous stirring for 1 h at 40 °C. After the complete reduction of Au³⁺ to Au° on the surface of α-Fe₂O₃-CeO₂ NC, the solid samples were centrifuged at 4000 rpm for 10 min and rinsed multiple times with deionized water and ethanol. Subsequently, the sample was further dried at 70 °C for 24 h. Eventually, α-Fe₂O₃-CeO₂@Au NC was collected and kept for further study. The schematic representation for the synthesis of α-Fe₂O₃, CeO₂, α-Fe₂O₃-CeO₂, and α-Fe₂O₃-CeO₂@Au NPs is illustrated in Figure 1.

2.3. Characterization

The surface morphology and elemental analysis of the prepared NPs were investigated by scanning electron microscopy (SEM) coupled with energy-dispersive X-ray spectrometry (EDS) (EDS Bruker, FEI Nova NanoSEM 450, Billerica, MA, USA). The determination of the sizes and shapes of the particles was carried out through the use of a transmission electron microscope (TEM) with Technai 200 model, which was produced by FEI and is based in Pleasanton, CA, USA. Fourier-transformed infrared spectra (FTIR) were also recorded in the range of 400–4000 cm⁻¹ on Bruker Alpha-II spectrometer, Billerica, MA, USA, to verify the essential function of functional groups in stabilizing, capping, and reducing during NP formation. Furthermore, Bruker D2 Phaser diffractometer, Billerica, MA, USA, was used to perform X-ray diffraction (XRD) analysis. The employed radiation was Cu Kα (1.5412 Å) over the 2θ range of 10–90°, with a step size of 0.002°, for the identification of photocatalysts’ phases. X-ray photoelectron spectroscopy (XPS) was carried out to analyze the chemical and elemental composition of synthesized materials using Kratos Analytical AMICUS instrument, Manchester, United Kingdom, outfitted with a dual Al-Mg anode and an achromatic X-ray source (1468.6 eV), with a step of 0.1 eV. UV-vis DRS analysis was performed using Thermo Scientific Evolution 220 UV–visible spectrophotometer, Waltham, MA, USA, to acquire the absorption spectra of the as-prepared samples over a 200–900 nm range. The photoluminescence (PL) spectral analysis was performed using a fluorescence spectrophotometer LS 45, Perkin Elmer, Waltham, MA, USA).

2.4. Photocatalytic Activity

In this work, RB dye was utilized as the standard pollutant to examine the photocatalytic effectiveness of α-Fe₂O₃-CeO₂@Au NC. The dye was degraded in an aqueous media (deionized water) using a quartz photocatalytic double-layer reactor in the presence of visible irradiation. Before conducting photocatalytic degradation experiments, preliminary studies were carried out to determine the optimal concentration of the dye and a catalyst for their specific system. Each experiment used 50 mg L⁻¹ of the photocatalyst, and an aqueous solution of RB (20 ppm) was used to disperse the sample in the dark for 30 min before irradiation to achieve adsorption–desorption equilibrium [49]. The reaction was conducted at a neutral pH of 7 [71,72] while maintaining the temperature at 30 °C using a circulating water-cooling system [73].

The photocatalyst was subsequently illuminated with visible light using a Tungsten lamp (400 W) under constant stirring. The concentration of the aqueous solution was determined using a UV-vis spectrophotometer at λ_max of 562 nm to monitor the decrease in the RB concentration over time. The α-Fe₂O₃-CeO₂@Au photocatalyst was then magnetically separated, washed, and dried for subsequent reactions to assess its stability and reusability. The photocatalytic experiments were optimized by varying photocatalyst and dye concentrations.

The data were also used to study the kinetic behavior of RB dye in the presence of the photocatalyst, and the rate constant (k) was calculated using Equation (1). The apparent rate constant of the reaction (k_app) in min⁻¹ was acquired from the slope of the ln A₀/A_t Vs time plot.

$$\ln \left(\frac{{\mathrm{A}}_0}{{\mathrm{A}}_{\mathrm{t}}}\right)={\mathrm{k}}_{\mathrm{app}} ·\mathrm{t}$$

(1)

Additionally, experiments were conducted to examine the effect of different free radicals on the photocatalytic process using various scavengers, including ethylenediaminetetraacetic acid (EDTA), tert-butyl alcohol (TBA), and benzoquinone (BQ), which were used as scavenging agents for h⁺, ^•OH, and ^•${\mathrm{O}}_2^{-}$, respectively [49,74,75].

3. Results and Discussion

3.1. FTIR Spectroscopy Analysis

The synthesized NPs, α-Fe₂O₃, CeO₂, α-Fe₂O₃-CeO_2,, and α-Fe₂O₃-CeO₂@Au, were explored using the FTIR spectroscopic analysis, as shown in Figure 2. In general, the spectra did not substantially differ with the deposition of the Au NPs. The peaks of all samples at about 1630 cm⁻¹ could be explained by the stretching and bending vibrations of O–H bonds in the absorbed molecular water. On the contrary, the stretching of O–H bonds in the hydroxyl groups of the capped ascorbic acid appeared as a broadband at around 3400 cm⁻¹ [76]. Furthermore, the peaks located at 2800–2900 cm⁻¹ could be correlated with the asymmetric and symmetric stretching of CH₂ in the methylene chains of amine, confirming that CTAB is coupled with the photocatalysts’ surfaces [77]. Meanwhile, the bands corresponding to the C–H bending of the methyl group appeared at approximately 1380 and 1460 cm⁻¹ [76,78]. In addition, FTIR spectra at 1050–1160 cm⁻¹ are associated with the C–O stretching vibration peaks that might refer to ester groups of ascorbic acid [79]. The observed peak intensities within the range of 1450–1700 cm⁻¹, 1462 cm⁻¹, and 1000–1250 cm⁻¹ were attributed to the presence of C=O, C–H stretching vibrations, and C–N stretching due to phenolic acids, terpenoid-phenols, and aliphatic amines, respectively [80,81].

The prominent peaks at approximately 550 and 460 cm⁻¹ were assigned to the bending and stretching vibrations of the Fe–O bonds [82], whereas the strong absorption bands at ~560 cm⁻¹ were attributed to Ce–O. These results were consistent with the XRD, XPS, and EDX results, indicating the formation of CeO₂ and α-Fe₂O₃ NPs [78]. However, the peaks associated with Au NPs did not emerge in the α-Fe₂O₃-CeO₂@Au spectra. Similar observations were reported in a previous study [76]. This occurrence could be due to either the prominent bands of the supporting nanoparticles masking the weaker bands of the gold nanoparticles or the low concentration of gold nanoparticles in the photocatalysts [33]. Furthermore, ascorbic absorption signals at ≈1750–1760, specifically attributed to the stretching of the carbonyl group in the γ-lactone ring, did not emerge [83]. Overall, the shift in peak intensities towards higher (redshift) wavelengths assured the deposition of Au NPs on the synthesized photocatalysts.

3.2. UV–Visible DRS Analysis

DRS UV-visible absorption spectroscopy was used to investigate the light absorption properties of the synthesized materials, as shown in Figure 3. The absorption edge of pure CeO₂ NPs, as shown in Figure 3a, was around 490 nm in the UV region, proving the transition of electrons in the formed Ce–O bond [84]. Moreover, a strong absorption peak emerged at about 380 nm for CeO_2, referring to the charge transfer from O 2P to Ce 4f orbitals [85]. On the other hand, the UV-visible spectrum of pure α-Fe₂O₃ NPs represented a strong absorption band in the visible-light region (400–720 nm), which might enhance the use of visible light for photocatalysis [86]. The absorption peaks close to 200 nm were attributed to the direct transition of the metal oxide NPs [87]. Moreover, the broad peak at around 550 nm corresponded to the process of pair excitation in α-Fe₂O₃ NPs, while the one at approximately 650 nm might belong to the d–d transition [88,89].

Compared to the individual metal oxide NPs, their NC exhibited a significant absorption ability under visible light irradiation. The strong absorption of visible light was linked to the significant amount of interfacial contact between CeO₂ and α-Fe₂O₃ NPs, forming a successful heterogeneous structure accountable for the UV–vis spectrum shifts and the bandgap width reduction [76]. Such an increase in the intensity might produce active electron–hole pairs, enhancing the photocatalytic efficiency [90,91]. In the case of Au-deposited α-Fe₂O₃-CeO₂, there was a slight gold band at around 520 nm, associated with the SPR effect; the band was not very prominent, and might overlap with the already existing broad peak of α-Fe₂O₃ at 530–570 nm. Furthermore, this could also be attributed to the comparatively minimal amount of Au NPs compared to the significant quantity of the support [92]. However, the deposition of Au NPs was confirmed with the results of XRD, XPS, and EDX [90]. The bandgaps were estimated for all samples from the carved edges of the Tauc plot by drawing (αhυ) against hυ. It was found that the bandgaps of α-Fe₂O_3, CeO₂, α-Fe₂O₃-CeO₂, and α-Fe₂O₃-CeO₂@Au were 1.92 eV, 2.80 eV, 1.80 eV, and 1.78 eV, respectively, as represented in Figure 3b–e. Consequently, the NC band gap energy (Eg) was lower than its corresponding metal-oxide-based NPs’, which was attributed to the formation of a heterojunction structure between α-Fe₂O₃ and CeO₂, which enhanced the surface electric charge and facilitated the transmission of photo-excited electrons [93]. Additionally, the difference in the bandgap of 0.02 eV for α-Fe₂O₃-CeO_2, after the gold deposition demonstrated the impact of the Au NPs on the photocatalyst’s surface. These findings further supported the photocatalysts’ catalytic activity in the visible-light region [94].

3.3. XRD Analysis

XRD analysis was used to investigate the crystal structure and purity of the as-prepared materials. Figure 4 shows the XRD patterns of α-Fe₂O₃, CeO₂, α-Fe₂O₃-CeO₂, and α-Fe₂O₃-CeO₂@Au nanoparticles. The XRD pattern of CeO₂ NPs shows characteristic peaks, observed at 28.6°, 33.2°, 47.6°, 56.6°, 59.4°, 69.8°, 76.9°, 79.3°, and 89.7°, corresponding to (111), (200), (220), (311), (222), (400), (331), (420), and (422) crystal planes, respectively. It can be inferred from the diffraction peaks that CeO₂ NPs possess a cubic fluorite structure and a face-centered cubic (FCC) structure and match well with the standard patterns of CeO₂ NPs (JCPDS-34-0394) [95]. The sharp diffraction peaks and the lack of additional peaks confirmed the high purity and crystallinity of the CeO₂ NPs [96]. The hematite phase (α-Fe₂O₃) was observed in the XRD pattern of pure Fe₂O₃ NPs. The NPs were crystalline in the rhombohedral lattice structure of α-Fe₂O₃ (JCPDS No. 24-0072), which had diffraction peaks at 2θ = 24.4°, 33.4°, 35.8°, 41.0°, 49.6°, 54.3°, 57.7°, 62.6°, 64.2°, 72.1°, 75.6°, 80.76°, 83.08°, 85.08°, and 89.10°, indexed as the (012), (104), (110), (113), (024), (116), (018), (214), (300), (1010), (217), (312), (0210), (134), and (126) planes [97]. The absence of secondary phases suggests that the produced material is pure and consists solely of the α-Fe₂O₃ single phase. The absence of any diffraction peak associated with the FeOOH phase suggests that the thermal annealing process entirely transformed the FeOOH into the α-Fe₂O₃ hematite phase. XRD investigations confirmed that the XRD patterns of α-Fe₂O₃-CeO₂ NC contain all the rhombohedral lattice structures of α-Fe₂O₃ and face-centered, cubic-structured CeO₂ NPs [98]. However, a slight reflection at around 35°, corresponding to the (110) plane of α-Fe₂O₃, was observed [96]. The majority of CeO₂ nanoparticles might be distributed on the surface of α-Fe₂O₃, effectively masking α-Fe₂O₃ and reducing its diffraction peaks in α-Fe₂O₃-CeO₂ NC, or the small amount of α-Fe₂O₃ NPs compared to CeO₂ in the nanocomposite [99].

Additionally, after gold deposition on the sample surface, additional peaks emerged at 38.2°, 44.4°, 64.6°, 77.6°, and 81.7° corresponding to the characteristic face-centered cubic (FCC) phase of gold (JCPDS no. 04-0784) [100]. The difference in their intensity reflected the strong supporting influence of gold dispersion on nanocomposite [101].

The average crystallite sizes (D) of the synthesized NPs and their NC were calculated based on the Debye Scherrer equation (Equation (2)):

$$\mathrm{D}=\frac{\mathrm{K}\mathsf{\lambda }}{\mathsf{\beta }\cos \mathsf{\theta }}$$

(2)

where λ is X-ray wavelength (0.154 nm), K represents the shape factor (0.9), β is the full width at half maximum of the diffraction band (FWHM), and θ is the Bragg diffraction angle [102]. The calculated crystallite sizes of CeO₂, α-Fe₂O₃, α-Fe₂O₃-CeO₂, and α-Fe₂O₃-CeO₂@Au were 9.43 nm, 6.94 nm, 7.92 nm, and 36.42 nm, respectively. The NC modified with Au NPs obtained larger crystallite sizes than its pure form and components.

3.4. XPS Analysis

The analysis using X-ray photoelectron spectroscopy (XPS) provides information about the chemical composition and oxidation state of elements (O, Ce, Fe, and Au) in α-Fe₂O₃-CeO₂@Au NC. Figure 5 provides high-resolution XPS spectra of Fe 2p, Ce 3d, O 1s, and Au 4f, along with the survey spectra of the NC. The survey spectrum in Figure 5a exhibits the peaks in binding energy associated with Ce 3d_3/2, Ce 3d_5/2, Fe 2p_3/2, Fe 2p_1/2, O 1s, C 1s, and Au 4f. There are two main peaks in the Fe 2p spectrum, located at 710.4 eV and 723.8 eV. These peaks correspond to Fe 2p_3/2 and Fe 2p_1/2, respectively, as well as two satellite peaks at 718.7 and 732.2 eV, which are attributed to Fe³ (Figure 5b). This confirmed the presence of the standard Hematite phase in the α-Fe₂O₃-CeO₂@Au NC [103]. The Ce 3d spectrum shows Ce³⁺ and Ce⁴⁺ oxidation states associated with Ce 3d_3/2 and Ce 3d_5/2, suggesting the existence of the CeO₂ NPs in the NC. The Ce 3d_5/2 peaks at 916.3, 897.9, and 888.4 eV mainly represent the vibration peaks of Ce⁴⁺, indicating that the primary oxidation state of CeO₂ is Ce(IV). On the other hand, peaks at 907.1 and 900.4 eV belong to Ce³⁺3d_3/2, and the ones at 882.0 eV refer to Ce³⁺3d_3/2 (Figure 5c) [104,105].

The O 1s spectrum demonstrated two broad peaks, with the lower binding energy peak (529.6 eV) assigned to the lattice oxygen of the metal oxide phase in CeO₂ NPs or Hematite and the higher binding energy peak (531.7 eV) corresponding to chemisorbed oxygen from hydroxyl (OH group) formed by oxygen (Figure 5d) [76,104]. The Au 4f XPS spectrum shows a doublet resulting from the spin-orbital splitting of 4f_7/2 and 4f_5/2 states, located at 83.6 eV and 87.2 eV, respectively, which is a typical characteristic of the presence of the Au NPs’ state (Figure 5e) [100]. Two peaks were visible on the C 1s spectrum (Figure S1): one at 285 eV and the other at 288.4 eV. The first peak indicated sp²-hybridized carbon (C–C), while the second peak indicated O–C=O [104,106]. The XPS results confirmed the successful incorporation of the Hematite NPs into the CeO₂ NPs and the Au NPs, which aligns with the XRD and EDX results.

3.5. TEM Analysis

The surface morphology and the size of the α-Fe₂O₃, CeO₂, α-Fe₂O₃-CeO₂, and α-Fe₂O₃-CeO₂@Au NPs were investigated by TEM analysis, as depicted in Figure 6. The α-Fe₂O₃ NPs exhibited an irregular spherical and mixed-morphology nanostructure with an average particle size of approximately 6 nm, with some particles appearing larger due to agglomeration (Figure 6a). CeO₂ NPs displayed a fluorite structure, manifested as agglomerates of fine nanocrystallites with an average size of approximately 7 nm (Figure 6b). TEM images of α-Fe₂O₃-CeO₂ nanocomposites (Figure 6c) revealed the formation of large agglomerations of semi-spherical structures consisting of fine and large nanoparticles with a diameter of approximately 7 nm. These findings confirmed the formation of a heterostructure between α-Fe₂O₃ and CeO₂. Furthermore, the deposition of Au NPs on the nanocomposite surface did not disrupt the α-Fe₂O₃-CeO₂ nanostructure (Figure 6d), which indicates the successful integration of spherical AuNPs with the heterojunction to produce α-Fe₂O₃-CeO₂@Au, which might hinder the recombination of e⁻–h⁺ pairs and enhance their photocatalytic efficiency [40].

3.6. SEM, Mapping, and EDX Analysis

The morphology and microstructure of α-Fe₂O₃, CeO₂, α-Fe₂O₃-CeO₂, and α-Fe₂O₃-CeO₂@Au were measured using SEM analysis, as represented in Figure 7. In Figure 7a, Hematite exhibits the aggregation of semi-spherical nanoparticles with a high-crystallinity nanostructure. Figure 7b displays CeO₂ NPs, which appear as an aggregation of semi-spherical fine nanoparticles. Meanwhile, the α-Fe₂O₃-CeO₂ nanocomposite exhibits the highest level of particle aggregation and showcases coral-like nanostructures with numerous vacancy defects on its surface. However, the original nanostructures of α-Fe₂O₃ NPs and CeO₂ NPs remained intact in the α-Fe₂O₃-CeO₂ NC, confirming the formation of a heterostructure between the two components (Figure 7c). Additionally, modifying α-Fe₂O₃ with CeO₂ NPs and Au NPs resulted in increased raggedness and surface rigidity (Figure 7d), while reducing particle aggregation. Consequently, the porosity of the surface increased, which facilitated greater contact between the active reactants. It is worth noting that this porous structure amplifies the surface area and creates additional pathways for charge carrier transfer, ultimately improving its photocatalytic activity [107].

Elemental mapping was utilized to confirm the steady distribution of elements on the α-Fe₂O₃-CeO₂@Au nanocomposite. The images (Figure S2b–e) demonstrated the homogeneous distribution of the primary constituent elements, namely Fe, Ce, and O, and the Au deposited on its surface. The nanocomposite’s elemental composition was further investigated using energy-dispersive X-ray (EDX) spectra (Figure S2f). The absence of impurities and the presence of strong peaks attributed to Fe, Ce, O, and Au in the nanocomposite confirmed its successful formation. The spectrum indicated that the nanocomposite contained a significant amount of Fe, Ce, O, and Au. These findings were in line with the findings of the XRD and XPS tests. Element mapping indicates the uniform distribution of Fe, Ce, O, and Au species throughout the nanoparticle.

3.7. PL Analysis

Efficiently separating e⁻–h⁺ pairs is crucial in photocatalytic applications. Therefore, photoluminescence (PL) spectra were performed to assess the effectiveness of separating the photogenerated electrons and holes of α-Fe₂O₃-CeO₂ and α-Fe₂O₃-CeO₂@Au NCs. Figure S3 revealed that the α-Fe₂O₃-CeO₂ exhibited a prominent peak at 430 nm, attributed to the recombination of rapidly generated photoelectrons and holes. The intensity of the peak in α-Fe₂O₃-CeO₂@Au was diminished, likely due to the formation of a heterojunction between α-Fe₂O₃-CeO₂ and Au NPs, which can serve as an efficient electron sink. This suggests that the recombination of photogenerated electrons and holes was suppressed [90].

3.8. Photocatalytic Activity

Based on the bandgap and PL results, α-Fe₂O₃-CeO₂@Au NC was utilized for the efficient photodegradation of RB dye under the influence of visible-light irradiation, as shown in Figure 8. The desired concentrations of the photocatalyst and aqueous dye solution were stirred in the dark for 30 min at a neutral pH before being exposed to visible-light irradiation to achieve adsorption–desorption equilibrium. Under dark conditions, no significant degradation of RB was observed, with a high absorption peak at 546 nm. However, the spectrum decreased gradually in the presence of visible light. Figure 8 (Insert) displayed that the degradation rate gradually increased with increasing irradiation time, reaching 88.9% within 120 min. The increase in the photodegradation rate with a longer exposure time is caused by the heterojunction formed between α-Fe₂O₃ and CeO₂. This junction promotes the transfer of charge between the semiconductors. The SPR effect of Au NP_S improves the absorption of visible light by the heterojunction, resulting in the generation of more photoelectron–hole pairs. Additionally, the Au NPs can trap electrons and enhance the separation of electron–hole pairs, leading to the generation of more hydroxyl radicals. These radicals are responsible for degrading the dye [40,108].

Furthermore, Table S1 compares the present photocatalytic activity of α-Fe₂O₃-CeO₂@Au NC with recent studies using Fe₂O₃-based nanoparticles under light irradiation [109,110,111,112,113,114,115,116,117,118].

3.9. Effect of Various Photocatalyst Concentrations

Optimizing the concentration of the α-Fe₂O₃-CeO₂@Au photocatalyst for the degradation of RB at a fixed pH, time, and temperature was the objective of a set of experiments that were carried out. The experiments involved different concentrations of the photocatalyst (20, 30, 40, 50 mg/L), as shown in Figure 9a. The results showed that as the concentration increased from 20 to 50 mg/L, the constant rate rose dramatically, from 0.00895 min⁻¹ to 0.01789 min⁻¹. The highest degradation rate was achieved at a 50 mg/L catalyst concentration. With the increase in catalytic concentration, the availability of active sites increased, resulting in the rapid adsorption of dye molecules as compared to comparatively minor concentrations, thus reducing turbidity to opacity and enhancing the scattering of light radiation in the mixture solution. As a result, the passage of irradiation was enhanced, causing higher numeric values of photodegradation at 50 mg/L. This fact can also be attributed to the increased density of active sites and absorbed photons generating free hydroxy radicals (^•OH) on the photocatalyst surface at a comparatively higher catalyst concentration, which led to the high photocatalytic degradation rate [119]. Therefore, 50 mg/L was selected as the optimum α-Fe₂O₃-CeO₂@Au concentration for further experiments for recovery cycle runs.

3.10. Effect of Various Dye Initial Concentrations

The influence of the initial concentration of RB on its photodegradation was examined using the α-Fe₂O₃-CeO₂@Au NC. The degradation rate was measured at a fixed amount of photocatalyst (50 mg/L), a neutral pH, and a temperature of 30 °C for 120 min. As the concentration of RB initially increased from 20 ppm to 50 ppm, the rate constant of the RB degradation declined from 0.01789 min⁻¹ to 0.00851 min⁻¹ (Figure 9b).

During the photodegradation process, increasing the initial concentrations of RB molecules promoted the interaction between active sites of α-Fe₂O₃-CeO₂@Au NC and RB molecules onto the surface, leading to enhanced turbidity and opacity. At elevated initial concentrations of RB molecules, the creation of electron–hole pairs decreased when exposed to light radiation with a constant catalytic concentration. Consequently, photo-degradation efficiency is reduced. Overall, the photodegradation process involves the generation of reactive oxidation species (ROS) of ^•OH, h⁺, ^•${\mathrm{O}}_2^{-}$ and e⁻. Hence, a decline in ROS with increased RB concentrations at a fixed concentration of α-Fe₂O₃-CeO₂@Au NC was observed in this study. Additionally, the catalysts’ surface quickly became saturated by dye molecules, leading to a shortage of active sites on which hydroxyl ions could adsorb. This results in a lower ^•OH concentration, which, in turn, reduces the photodegradation efficiency, as the ^•OH’s contribution to the photo-degradation of dye molecules is significant [119]. Thus, we concluded that the comparatively lower concentration of RB, i.e., 20 ppm, is an optimum concentration for photodegradation in the presence of the α-Fe₂O₃-CeO₂@Au NC and light radiation. Table S2 presents the values of k (pseudo-first-order reaction rate constants) and half-lives for the photocatalytic degradation of RB using the α-Fe₂O₃-CeO₂@Au photocatalyst under different catalyst concentrations and initial concentrations.

3.11. Reusability Study

The collection and reusability of photocatalysts are crucial, and one highly effective approach to achieve this involves the utilization of magnetic materials [51]. In this study, α-Fe₂O₃ and CeO₂ were combined to form the α-Fe₂O₃-CeO₂@Au heterojunction, as hematite is the most suitable magnetic material due to its high magnetic properties, saturation magnetization, and coercivity [90]. The α-Fe₂O₃-CeO₂@Au NC showed excellent magnetic recyclability, as it could be rapidly isolated from the solution through the application of an external magnetic field. The photocatalyst stability was evaluated through 1–4 cycle experiments under visible-light irradiation (Figure 9c). After four cycles, the degradation of RB was observed to drop from 89% to 64% with less than a 25% decrease in its initial activity. After conducting four photocatalytic tests, the photodegradability of the catalyst decreased in comparison to the original photocatalyst owing to the degradation of the photocatalyst over time, which can cause a decrease in the efficiency the generation of reactive species. Additionally, the adsorption of the RB dye or other contaminants on the surface during the recycling process hindered the active sites of the photocatalyst’s surface and reduced its photocatalytic activity [120]. This investigation indicates that the present photocatalyst can lead to the efficient photocatalytic degradation of RB dye due to its high photostability and confirms its reasonable catalytic reusability. To enhance the stability and reusability of the photocatalyst, it may be necessary to optimize or modify it without compromising its performance.

3.12. Effect of Radical Scavengers

To better understand the RB photocatalytic degradation process using α-Fe₂O₃-CeO₂@Au NC, a series of experiments were performed to deactivate potential reactive species by employing different scavengers for the purpose of radical scavenging (Figure 9d).

The experiments were performed under optimal conditions, which included an initial dye concentration of 20 ppm, a catalyst dose of 50 mg/L, a neutral pH, and an irradiation time of 120 min at 30 °C. To scavenge h⁺, ^•OH, and ^•${\mathrm{O}}_2^{-}$, three chemical scavengers, namely ethylenediaminetetraacetic acid (EDTA; 0.5 mM), tert-butyl alcohol (TBA; 0.5 mM), and benzoquinone (BQ; 0.5 mM), were used. There was also a control group without free radical scavengers. The results showed that EDTA, TBA, and BQ inhibited the reaction rate by approximately 40%, 46%, and 14%, respectively. The degradation effectiveness of RB exhibited a notable decrease upon the addition of TBA and EDTA. These findings indicated that ^•OH and h⁺ were the main reactive species contributing to the reaction process. This could be attributed to the hetero-junction formed between α-Fe₂O₃ and CeO₂, and the presence of deposited gold nanoparticles, which enhanced the charge transfer and, hence, electron–hole pair separation. On the other hand, the addition of BQ led to only a minor reduction in degradation efficiency, indicating that ^•${\mathrm{O}}_2^{-}$ did not play a significant role in the photocatalytic degradation despite its known involvement in the chain of heterogeneous/homogenous photocatalytic reactions [121]. This could be due to the consumption of superoxide radicals by other compounds present in the system, thereby reducing their overall impact on the degradation process [122]. Overall, the results suggested that ^•OH and h⁺ were the primary reactive species involved in RB degradation, while ^•${\mathrm{O}}_2^{-}$ had a low impact on the reaction process.

3.13. The Mechanism for Photodegrading RB Dye Using α-Fe₂O₃-CeO₂@Au NC

The photocatalytic technique has several distinguishing characteristics, such as specific bandgaps, as well as a specific surface area, morphology, reusability, and stability. In this study, a proposed photodegradation mechanism for RB dye by α-Fe₂O₃-CeO₂@Au heterojunction was presented based on the experimental results, as illustrated schematically in Figure 10. Upon exposure to visible light, photogenerated electrons in the valence bands (VB) of both CeO₂ and α-Fe₂O₃ were excited to their conduction bands (CB), leading to the generation of photogenerated holes in the valence bands. The CB of α-Fe₂O₃ (0.41 eV) is less negative than that of CeO₂ (−0.34). Therefore, the photogenerated electrons moved from CeO₂ to α-Fe₂O₃. Similarly, the VB of α-Fe₂O₃ is less positive (2.33 eV) than that of CeO₂ (2.46 eV). Hence, the photogenerated holes in CeO₂ also moved to α-Fe₂O₃. The creation of the Schottky barrier at the nanocomposite interface facilitated the charge transfer between the semiconductors [123]. In addition, under light irradiation, the surface plasmonic resonance (SPR) effect enhanced the efficiency of light absorption and resulted in the generation of additional photogenerated electron–hole pairs in the α-Fe₂O₃-CeO₂@Au nanocomposite [57,124]. However, the electrons accumulated on the CB of α-Fe₂O₃ cannot convert dissolved oxygen into ^•${\mathrm{O}}_2^{-}$ because the CB potential of α-Fe₂O₃ (0.41 eV) is less negative than the electrode potential of O₂/^•${\mathrm{O}}_2^{-}$ (−0.33 eV vs. NHE). The SPR effect of Au NPs improved their ability to capture electrons, which assisted in separating the photogenerated electron–hole pairs and reduced the probability of recombination. Hence, the accumulated electrons on the CB of α-Fe₂O₃ can be easily trapped by Au NPs and then captured by the adsorbed O₂ to generate ^•${\mathrm{O}}_2^{-}$. This superoxide anion undergoes an additional reaction with water, leading to the generation of the highly reactive oxidizing agent, hydroxyl radical [122], which is consistent with the scavenger’s findings. These transfers occurred much faster than the recombination of photogenerated electrons and holes in α-Fe₂O₃ [125]. The holes on VB of α-Fe₂O₃ reacted with OH⁻ to produce ^•OH, since the VB of α-Fe₂O₃ is more positive than the electrode potential of ^•OH/H₂O (1.99 eV vs. NHE) [126]. The reactive oxidation species, including ^•${\mathrm{O}}_2^{-}$, ^•OH, and h⁺, interacted with the RB molecules, contributing to their degradation [127]. In summary, the degradation process could be described by the following reaction:

α-Fe₂O₃ + hv → e⁻ (α-Fe₂O₃) + h⁺ (α-Fe₂O₃)

(3)

CeO₂ + hv → e⁻ (CeO₂) + h⁺ (CeO₂)

(4)

e⁻ (CeO₂) CB → e⁻ (α-Fe₂O₃)

(5)

h⁺ (CeO₂) VB → h⁺ (α-Fe₂O₃)

(6)

e⁻ (α-Fe₂O₃) → e⁻ (Au°)

(7)

$${\mathrm{e}}^{-} (\mathrm{Au}°)+{\mathrm{O}}_2 \to { }^{•}{\mathrm{O}}_2^{-}$$

(8)

$${\mathrm{H}}_2{\mathrm{O}}_2+{ }^{•}{\mathrm{O}}_2^{-} \to { }^{•}\mathrm{OH}$$

(9)

h⁺ + (α-Fe₂O₃) + OH⁻ → ^•OH

(10)

$$\mathrm{RB} \mathrm{dye}+{\mathrm{h}}^{+}+{ }^{•}{\mathrm{O}}_2^{-}+{ }^{•}\mathrm{OH} \to {\mathrm{H}}_2\mathrm{O}+{\mathrm{CO}}_2$$

(11)

The band position of the as-synthesized samples was determined using empirical Equations (12) and (13).

E_CB = χ − E_e − 0.5E_g

(12)

E_VB = E_CB + E_g

(13)

where χ is the electronegativity of the material and the χ values of CeO₂ and α-Fe₂O₃ were estimated previously to be 5.56 [128,129,130]; E_e is the energy of the free electrons on the hydrogen scale (4.5 eV); E_g represents the band gap of the material; E_CB stands for the conduction band potential; and E_VB denotes the valence band potential. The values of χ, E_g, E_CB, and E_VB for the synthesized samples are provided in Table S3.

4. Conclusions

A novel nanocomposite (NC) α-Fe₂O₃-CeO₂@Au was developed to efficiently eliminate Rose Bengal (RB) dye from wastewater using a photocatalytic technique under visible light. The NC was synthesized with stabilizing or capping agents, and the surface was coated with gold nanoparticles. The as-synthesized samples underwent characterizations of their structural, morphological, and optical properties. The study evaluated the effectiveness of α-Fe₂O₃-CeO₂@Au photocatalyst concentrations, initial dye concentration, and radical scavengers on the photocatalytic degradation of RB. The nanocomposite α-Fe₂O₃-CeO₂@Au (50 mg L⁻¹) achieved 88.9% for the removal of RB dye (20 ppm) in 120 min under visible-light irradiation at a pH of 7 and 30 °C. The results of the radical scavenging experiments suggested that holes (h⁺) and hydroxyl radicals (^•OH) were the main reactive species involved in the Rose Bengal degradation. The study highlights the potential of the nanocomposite for treating wastewater, especially for removing organic dyes. This composite can be easily separated using a magnetic field, offering an efficient method for catalyst recovery and suggesting low operating expenses. However, the effectiveness of the composite decreases with repeated use, which raises sustainability concerns. To ensure the long-term sustainability of the photocatalytic treatment process, it is crucial to address the issues of catalyst stability and reusability.